US 7422871 B2
Disclosed are methods of detecting uropathogenic E. coli genes that are differentially expressed in response to D-serine. Also disclosed are methods of characterizing bacterial isolates from clinical samples based on the ability to metabolize D-serine.
1. A method of characterizing an E. coli strain isolated from a clinical sample comprising testing the strain for the ability to grow in the presence of D-serine.
2. The method of
3. The method of
4. The method of
5. The method of
6. The method of
7. The method of
This application is a divisional of U.S. patent application Ser. No. 10/117,417, filed Apr. 5, 2002, which claims priority to U.S. Provisional Application Ser. No. 60/281,859, filed Apr. 5, 2001, which is incorporated herein in its entirety.
This invention was made with United States government support awarded by the following agencies: NIH AI39000. The United States has certain rights in this invention.
Urinary tract infections (UTIs) are responsible for an estimated 10 million outpatient visits and more than one million hospitalizations each year in the United States. Annual health care costs attributable to UTIs exceed one billion dollars. Approximately 50% of women will have at least one UTI during their lives, and many of these women experience recurrent UTI.
Uropathogenic Eschericia coli (UPEC), by far the most common uropathogen, is the causal agent in greater than 85% of uncomplicated UTIs. The organisms are also known to cause hospital acquired infections associated with injury, surgical procedures, or catheterization. UPEC present a complex ecological case among the pathogenic E. coli. UPEC can colonize the human intestinal tract without causing disease, yet they can sequentially colonize the perineum, urethra, bladder and kidney. Known E. coli virulence factors that contribute to UTIs are fimbriae (e.g. type 1 fimbriae and Pap fimbriae), iron acquisition systems, hemolysin, cytotoxic necrotizing factor-1, and the Sat autotransporter. Environmental cues used by UPEC for expression of genes needed for colonization of the bladder and kidney have not been described in the literature.
Urinary tract infections in humans are generally treated with antibiotic therapy. Treatment of UTIs may be complicated by strains that are resistant to antibiotics, particularly to those antibiotics in common use. The identification of new UPEC-specified pathogenicity factors will facilitate the development of new methods of treating or preventing UTIs.
In one aspect, the present invention provides a method of detecting uropathogenic E. coli nucleotide sequences differentially expressed in the presence or absence of D-serine comprising providing a library comprising a plurality of transposon mutants of a uropathogenic E. coli strain, the mutants comprising a transcriptional fusion comprising a transcriptional regulation sequence operably connected to a sequence encoding a detectable protein; growing the mutants in the presence or absence of D-serine; identifying mutants having increased or decreased expression of the transcriptional fusion in the presence or absence of D-serine by comparing the relative levels of the detectable protein of the mutants grown in the presence or absence of D-serine; identifying the insertion site of the transcriptional fusion in the identified transposon mutants; and correlating the insertion site with an E. coli nucleotide sequence.
In another aspect, the present invention provides a method of identifying proteins differentially expressed in a wild-type uropathogenic E. coli strain and a uropathogenic E. coli dsdCXA locus mutant, the mutant having reduced expression of one or more proteins selected from the group consisting of DsdA, DsdC, and DsdX, the method comprising comparing proteins isolated from the wild-type uropathogenic E. coli strain and proteins isolated from the uropathogenic E. coli dsdCXA locus mutant; and identifying proteins from the dsdCXA mutant having increased or decreased level of expression relative to expression of the corresponding proteins in the wild-type uropathogenic E. coli strain.
In other aspects, the invention includes a method of identifying a uropathogenic E. coli polynucleotide sequence that binds to DsdC protein comprising contacting an E. coli polynucleotide sequence with DsdC protein and detecting binding of the polynucleotide sequence to the protein.
In another aspect, the present invention provides a method of detecting genes from a uropathogenic E. coli strain that are differentially expressed in the presence or absence of D-serine comprising hybridizing a first set of labeled oligonucleotide probes with an array of oligomers, the oligomers comprising gene sequences from the uropathogenic E. coli strain, wherein the first set of labeled oligonucleotide probes is made by reverse transcription of RNA isolated from uropathogenic E. coli grown in the presence of D-serine; hybridizing a second set of labeled oligonucleotide probes with an array of oligomers identical to the aforementioned array, wherein the second set of labeled oligonucleotide probes is made by reverse transcription of RNA isolated from uropathogenic E. coli grown in the absence of D-serine; comparing hybridization of labeled oligonucleotide probes to identify oligomers having differential hybridization to the first and second sets of oligonucleotide probes; and identifying genes comprising the sequences of the identified oligomers.
In yet another aspect, the present invention provides a method of detecting proteins differentially expressed in a uropathogenic E. coli strain in response to D-serine, comprising providing a first culture of the uropathogenic E. coli strain grown in the presence of D-serine and a second culture of the uropathogenic E. coli strain grown in the absence of D-serine; comparing proteins isolated from the two cultures, and identifying proteins from E. coli grown in the presence of D-serine that are increased or decreased relative to the corresponding proteins in the uropathogenic E. coli strain grown in the absence of D-serine.
The present invention also provides a method of characterizing bacterial strains isolated from clinical samples comprising testing the strain for the ability to grow in the presence of D-serine.
In another aspect, the present invention provides urine dipstick comprising a solid support, a polypeptide comprising D-serine deaminase and an indicator responsive to ammonia, wherein the polypeptide and indicator are coimmobilized on the support.
The sequence of the uropathogenic E. coli strain CFT073 genome was compared with the sequences of the genomes of E. coli laboratory K-12 strain MG1655 and enterohemorrhagic E. coli (EHEC) 0157 strain EDL933. A sequence found in uropathogenic E. coli strain CFT073 but not in E. coli K-12 or EHEC 0157 was identified in the 53 minute region of the K-12 genetic map. The sequence includes the D-serine tolerance (dsdCXA) locus, which is involved in overcoming D-serine toxicity, and is presented in SEQ ID NO:13.
The dsd locus includes coding sequences for a positive transcriptional activator (DsdC), a putative permease (DsdX), and a D-serine deaminase (DsdA), which catalyzes the hydrolysis of D-serine to give pyruvate and ammonia. The 53 minute region includes two putative recombinases unique to UPEC, which are designated ipuA and ipuB. The DNA sequences of dsdC, dsdX dsdA, ipuA and ipuB are provided in SEQ ID NO:1, SEQ ID NO:3, SEQ ID NO:5, SEQ ID NO:7, and SEQ ID NO:9, respectively, and the deduced amino acid sequences of their putative translation products are shown in SEQ ID NO:2, SEQ ID NO:4, SEQ ID NO:6, SEQ ID NO:8, and SEQ ID NO:10, respectively. With reference to the SEQ ID NO:13, bases 3507-4844 correspond to SEQ ID NO:3; bases 4862-6190 correspond to SEQ ID NO:5; bases 1378-1947 correspond to SEQ ID NO:9; bases 3295-2342 correspond to the reverse complement of SEQ ID NO:1; and bases 623-7 correspond to the reverse complement of SEQ ID NO:7.
Based on their deduced amino acid sequences, the proteins putatively encoded by ipuA and ipuB are homologs of the recombinase fimB, which suggests that the proteins may function to regulate expression of the dsd operon in a manner similar to the regulation of type 1 fimbriae by FimB and FimE.
The polypeptide putatively encoded by dsdC is a member of the LysR family of transcriptional regulators. The LysR-type regulators are very common in bacteria and control the expression of genes associated with a wide range of cellular processes, including many diverse virulence factors. Members of the LysR family of transcriptional regulators usually require the presence of a specific coinducer to activate transcription. The polypeptides are about 270-330 amino acid residues in length and exhibit sequence similarity over about 270 amino acid residues. The strongest conservation occurs in the N-terminal 66 amino acid residues, which has a helix-turn-helix motif. The LysR-type regulators bind to DNA as dimers or tetramers and activate gene transcription, possibly by introducing a bend in the DNA helix.
Until recently, it was considered axiomatic that naturally occurring D-amino acids occurred only in bacteria, which contain D-amino acids in their cell walls. Bruckner et al. reported the presence of D-amino acids, including D-serine, in the urine from seven human subjects (Bruckner, et al., Amino Acids 1994. 6:205-211). Of the D-amino acids in urine, D-serine was present in the highest concentration (30 to 380 μMol/l). The percentage of D-serine relative to total (D-plus L-) serine ranged from 19% to 56%. Bruckner et al. speculated that D-amino acids in urine are derived from dietary sources (e.g., Lactobacillus in food).
It was recently reported that L-serine is converted to D-serine in mammalian brain by direct, enzyme-catalyzed racemization (Wolosker et al., PNAS 96:721-755, 1999). The highest concentrations of D-serine are found in regions of the brain having N-methyl-D-aspartate (NMDA) receptors for the neurotransmitter glutamate. D-serine functions as a potent agonist at the NMDA-glycine site and with glutamate can activate the NMDA receptor (Mothet et al. PNAS 97:4926-4931, 2000).
Uropathogenic E. coli are able to grow in media or urine in the presence of D-serine, whereas E. coli strains that lack the dsdCXA genes are unable to grow or grow more slowly than UPEC strains. As shown in the Examples below, three collections of E. coli were examined for the ability to utilize media having D-serine as the sole source of carbon and nitrogen. The results showed that 49 of 60 tested UPEC strains were able to use D-serine as the sole source of carbon and nitrogen. In contrast, just four of 74 tested hemorrhagic colitis and infantile diarrheal E. coli strains were able to use D-serine as the sole source of carbon and nitrogen.
As used herein, a “wild-type UPEC strain” is a strain of uropathogenic E. coli having the ability to use D-serine as the sole source of carbon and nitrogen.
To evaluate the role of the dsd operon in UPEC pathogenicity, a mutant in wild-type UPEC strain CFT073 with a deletion in the dsdA gene was constructed as described in the Examples. The dsdA mutant was evaluated for its ability to establish infection in a murine model of ascending urinary tract infection. Surprisingly, the mutant strain was present in the bladders and kidneys 48 hours postinfection at levels of about 300 times that of the wild-type strain. This result runs counter to our expectation that a knockout of this putative virulence gene would be less effective than wild-type CFT073 in establishing an infection. Complementation of the dsdA mutant with DsdA supplied in trans resulted in recovery similar to wild type.
The superior ability of the dsdA mutant over wild-type CFT073 to establish an infection in the murine coinfection experiments caused us to hypothesize that UPEC use D-serine as a signal to modulate growth and/or expression of virulence determinants. D-serine may act as an effector of the LysR-type transcriptional regulator, DsdC, to cause an increase in transcription. Production of DsdA would lower D-serine concentrations, which, if D-serine acts as an effector with DsdC, would reduce DsdC-regulated transcription. However, when concentrations of D-serine are relatively high, DsdC may remain active and cause the transcription of a myriad of genes for other virulence factors. It follows that people susceptible to recurrent UTIs caused by UPEC may produce urine with higher levels of D-serine than those who do not experience recurrent UTIs.
The identification of genes whose expression is regulated by D-serine, DsdC, or DsdC and D-serine will facilitate the design of drugs and methods of treating UTIs. Several approaches may be used to identify genes or proteins as targets for drug therapy.
One approach to identifying D-serine-regulated genes to create a library of transposon mutants by transposon-mediated mutagenesis, as described below in the Examples. Briefly, transposon mutagenesis is used to insert a gene encoding a detectable protein into the bacterial chromosome under the control of a transcriptional regulation sequence. Examples of suitable detectable proteins include, but are not limited to, luciferase and the lacZ gene product. Mutants expressing the detectable protein are identified and grown in the presence in or absence of D-serine. The relative expression of detectable protein in mutants grown in the presence and absence of D-serine is assessed, and those mutants exhibiting differential expression of the detectable protein in the presence and absence of D-serine are selected for further analysis.
Another approach to identifying genes whose expression is affected by DsdC or D-serine metabolism is to compare the proteins expressed by a wild-type UPEC strain grown in the presence or absence of D-serine, or expressed by a wild-type UPEC strain and a mutant of the strain comprising a mutation in dsdC, dsdA, or dsdX grown in the presence or absence of D-serine. Proteins are isolated from the bacterial cultures, separated by two-dimensional gel electrophoresis, and differential expression identified by comparing gels obtained from different bacterial strains or by the same bacterial strain grown in the presence or absence of D-serine. Proteins of interest may be further characterized to permit identification.
In another approach, RNA is isolated from a wild-type UPEC strain grown in the presence or absence of D-serine and subjected to reverse transcription to create labeled oligonucleotide probes, which in turn are hybridized to an oligomer array of E. coli gene sequences to identify sequences differentially expressed in the presence or absence of D-serine.
Using the methods described below in the Examples, several sequences differentially expressed in the presence of D-serine have been identified. Further identification and characterization of differentially expressed sequences is underway. Loss of function mutants for proteins identified as being of potential interest will be developed and evaluated for virulence using any suitable method. One method of evaluating virulence is an intracellular invasion assay, described in the Examples. Proteins that are correlated with virulence may be used to develop a vaccine against UPEC-mediated UTIs, to develop antibodies for use in treating UTIs by passive immunization, or to screen potential therapeutic agents. Conveniently, once a protein of interest is identified, the protein may be obtained by cloning the gene encoding the protein into an expression vector and obtaining expression in a suitable host cell.
The ability of certain UPEC strains to grow on D-serine will provide a simple means of differentiating E. coli isolates from urine or blood. The bacterial isolate may be tested for the ability to grow on or in medium containing D-serine as the sole carbon and nitrogen source. The bacterial isolate may be plated on a suitable medium comprising D-serine in a concentration effective to support growth of a UPEC strain capable of using D-serine as the sole source of carbon and nitrogen, suitably in the range of from about 100
Whether a correlation between UTI susceptibility and urinary D-serine levels exists may be determined by assaying D-serine in the urine of human populations having differential susceptibility to UTIs. If such a correlation exists, assaying D-serine may be useful in identifying individuals having this particular susceptibility. Unfortunately, D-serine assays are presently performed by HPLC and are prohibitively expensive for routine use.
The recognition that D-serine may play a role in establishing UTIs suggests the importance of having a simple assay for assaying D-serine levels. D-serine deaminase may be used to manufacture urine dipsticks and biosensors for quantifying D-serine. The D-serine deaminase may be coimmobilized on a suitable support with a suitable indicator that changes color in the presence of the ammonia generated by the enzymatic hydrolyis of D-serine. When the immobilized D-serine deaminase is exposed to a sample containing D-serine, the D-serine will be hydrolyzed to form pyruvate and ammonia, which will cause the indicator to change color, thus permitting the detection of the D-serine in the sample.
The D-serine deaminase used in urine dipsticks or biosensors may be obtained from any source. For example, D-serine deaminase may be obtained by subcloning a D-serine deaminase gene, such as the CFT073 dsdA gene (SEQ ID NO:6), into a suitable expression vector, introducing the vector into a suitable host cell, allowing expression of the protein, and recovering from the host cell D-serine deaminase substantially free of other proteins.
Suitable indicators include any compound that detectably changes in response to the hydrolysis of D-serine. For example, a suitable indicator may include an indicator that changes color in response to pH changes, which would occur with an increase in ammonia concentration. An example of such an indicator is phenol red.
The following non-limiting examples are intended to be purely illustrative.
Urosepsis E. coli strain CFT073 was originally isolated from the blood and urine of a woman admitted to the University of Maryland Medical System (Mobley et al., Infect. Immun., 1990, 58: 1281-1289). E. coli strain collections used in epidemiology experiments were a collection of urosepsis and urinary tract isolates (Arthur et al., Infect. Immun., 1989a, 57: 314-321; Arthur et al., Infect. Immun., 1989b, 57: 303-313; Arthur et al., Infect. Immun., 1990, 58: 471-479) and the DEC collection of hemorrhagic colitis and infantile diarrheal strains (Whittam et al., Infection and Immunity, 1993, 61: 1619-1629). Other strains used are described in Table 1: Bacterial strains and plasmids
To assess whether bacteria are able to utilize D-serine as the sole carbon and nitrogen source, bacteria were plated on a modified Minimal A Medium agar (15 g K2HPO4, 4.5 g KH2PO4, 1 g K2SO4, 0.5 g Na3C6H5O7.2H2O, 15 g agar and 1 ml of 1 M solution MgSO4.7H2O per liter) supplemented with 500 ug/ml of D-serine (Sigma) (Miller, Experiments in molecular genetics, 1972). For growth studies, bacteria were grown in MOPS minimal medium [3-(N-morpholino)propanesulfonic acid] supplemented with 10 μM thiamine and 0.4% glycerol (Neidhardt et al., J. Bacteriol., 1974, 119: 736-747). For other experiments, the media used included L broth (5 g of sodium chloride, 5 g of yeast extract and 10 g of tryptone per liter [Fisher Scientific]) and L agar (L broth containing 1.5% agar [Difco]). Human urine was provided by two different donors and filter sterilized prior to use in experiments. Mouse urine was collected from a group of eight mice and UV sterilized prior to use. Antibiotics were added as needed at the following concentrations: nalidixic acid and kanamycin at 50 μg/ml and chloraramphenicol at 20 μg/ml (Sigma).
Liquid cultures were grown at 37° degrees with aeration except for cultures used in mouse UTI experiments. In the mouse UTI experiments, each strain was inoculated into L-broth and grown without aeration for two days. Bacteria were then passaged into fresh medium two additional times for a total of 6 days growth.
To determine growth curves, single colonies were inoculated into 5 ml of the appropriate medium and grown overnight at 37° C. with aeration. Following the overnight incubation, cells were collected by centrifugation, washed in saline and resuspended in 0.5 ml of saline (8.5 g NaCl per liter). The bacteria were then added to pre-warmed tubes of the desired medium. The OD600 was adjusted to 0.03 with the appropriate medium. The cultures were grown at 37° C. with aeration and OD600 readings were taken at the designated time points and dilutions were plated to determine CFUs.
Restriction and DNA modification enzymes were purchased from either New England Biolabs or Promega Corporation. dsdA-specific primers for Polymerase Chain Reaction (PCR) were designed using the CFT073 genome sequence (R. A. Welch, unpublished). The dsdA primers used were 5′-GGATGGCGATGCTGCGTTG-3′ (SEQ ID NO:11) and 5′-CACAGGGGAAGGTGAGATATGC-3′ (IDT) (SEQ ID NO:12). The template was wild-type CFT073 genomic DNA prepared using the Wizard Genomic DNA Isolation Kit (Promega). The PCR amplification used the Expand High Fidelity PCR System (Roche Diagnostics) according to the manufacturer's protocol. The resulting PCR product was gel purified with the QIAQuick Gel Purification Kit (Qiagen) and ligated into pGEMt (Promega). The resulting dsda recombinant plasmid was examined by restriction endonuclease analysis. A dsdA deletion was constructed by digesting the dsdA plasmid with PmeI and EcoRI, which resulted in a 445 bp deletion. The digested plasmid was treated with T4 DNA polymerase in order to blunt the DNA ends. After gel purification, the plasmid was religated. An ApaI-SacI fragment containing dsda was subcloned into the suicide vector, pEP 185.2 (kind gift of Virginia Miller, Washington University). This construct was sequenced (ACGT, Incorporated) using T3 and T7 primers provided by ACGT. Conjugation was carried out as described by Kinder et al. (Kinder et al., Gene, 1993, 136: 271-275) with some modifications. The suicide plasmid construct was transformed into a S 17λpir background and then conjugated into a nalidixic acid resistant mutant of CFT073. Single-crossover chromosomal integrates of the recombinant plasmid were selected and D-cycloserine enrichment was used to select for a second crossover event that excised the suicide plasmid backbone from the CFT073 chromosome. Chromosomal DNA was isolated from the putative dsdA mutants and examined via PCR analysis using dsdA-specific primers to confirm the deletion. Phenotypic analysis of the putative mutants was carried out by inoculating colonies onto MOPS glycerol minimal and MOPS minimal D-serine agar media.
The dsdA gene, including its Shine and Delgarno sequence, was PCR amplified with primers (5′GCGCTGCAGCGTTATTAACGGCCTTTTGCCA GATATTGATTC) (SEQ ID NO:17) and (5′CGCGGATCCCGTACTAT GGAAAACGCTAAAA TGAATTCGC) (SEQ ID NO:18). The primers were designed with PstI and BamHI restriction sites at their respective 5′ ends. The PCR product was digested with PstI and BamHI enzymes, gel purified and ligated with pACYC177 previously digested with the same two restriction enzymes. The plasmid, designated pWAM2682, was used in subsequent complementation analysis.
After 6 days of static growth, the density of each bacterial strain was adjusted to an OD600 of 0.6. Equal numbers of wild-type CFT073 cells and dsdA mutant (nalidixic acid resistant background) were mixed and pelleted by centrifugation for 15 min at 7000 rpm. The pellet was resuspended in 500 μl of phosphate buffered saline (7.5 g NaCl, 0.2 g KCl, 0.2 g KH2PO4, 1.15 g Na2HPO4 per liter). Female CBA/JCrHsd mice (five to eight weeks of age, Harlan) were inoculated via trans-urethral catheterization with 50 μl of the prepared bacteria (approximately 1×109 organisms). After 48 h, the mice were sacrificed and their bladders and kidneys harvested. The organs were homogenized and dilutions plated on L agar without antibiotics. After overnight incubation, colonies were counted and then replica plated onto L agar containing nalidixic acid in order to determine the number of surviving dsdA mutants. The nalidixic acid resistant CFT073 background strain used in the construction of the dsdA mutant competes equally with the CFT073 nalidixic acid sensitive parent strain in the UTI model (Torres et al., Infect. Immun., 2001, 69: 6179-6185).
For the in vivo complementation studies, both the wild-type and dsdA mutant strains carried the complementing dsdA plasmid, pWAM2682. Both strains were grown as above with the addition of kanamycin in order to assure maintenance of the plasmids during the six day culture. Results were analyzed for statistical significance via paired Student's t test.
The wild-type and dsdA mutant strains were grown as described for the murine infection studies. The passaged cultures were diluted to an OD600 of 0.6 and CFUs determined by plate count. The assay was carried out in a 96 well U-bottomed microtiter plate. Serial two-fold dilutions of each culture (50 μl) were mixed with 50 μl of a 25% suspension of guinea pig erythrocytes in saline. The mixtures were incubated at room temperature and agglutination titers were determined after 1.5 h. In these experiments, CFT073 mutants locked ON and locked OFF for type 1 fimbriae expression (N. W. Gunther, unpublished) were used as hemagglutination controls (kind gift of N. W. Gunther and H. L. Mobley).
Wild-type and dsdA mutant strains grown overnight in human urine were inoculated into fresh urine and examined microscopically after 3 and 24 h of growth. The cell suspensions were air-dried on glass slides and stained with safranin. The cells were visualized using a Zeiss Axioplan IIi microscope and the images were collected digitally using OpenLabs 3.02 software (Improvision, Inc.).
A CFT073 lacZY deletion mutant was isolated using the λ red recombinase mutagenesis system of Datsenko and Wanner (Datsenko et al., Proc. Natl. Acad. Sci. USA, 2000, 97(12): 6640-6645), which is incorporated by reference herein. Briefly, PCR products were generated using primers with 36- to 50-nucleotide extensions that are homologous to the region adjacent to the lacZYA genes and template plasmid carrying an antibiotic resistance gene flanked by FLP recognition target sites. The PCR product was introduced into CFT073 carrying a Red expression plasmid via electroporation. After selection, the resistance gene was eliminated introduction of a helper plasmid which acts directly on the FLP recognition target sites flanking the resistance gene. The temperature sensitive Red and FLP helper plasmids are cured by growth at 37 degrees C.
MiniTn10lux transposon insertion mutants of CTF073 were constructed using miniTn10lux. E. coli strain KV330 (carrying the suicide plasmid pKV45 which harbors the mini-TN10 transposon, transposase and resolvase) was the donor strain for conjugal transfer into recipient host strain CFT073. Transconjugants were plated on LB agar supplemented with chloramphenicol (to select for the transposon) and nalidixic acid (to select for the host strain, CFT073). Individual isolates were assayed for luciferase expression in the absence of presence of D-serine.
MiniTn5lacZY transposon insertion mutants of a CFT073 lacZY deletion mutant were constructed as follows. E. coli S 17-1λpir containing the pREV 10 plasmid (this plasmid cannot be maintained in CFT073, but contains a MiniTn5 Km2 transposon with an added promoterless lacZY gene) was the donor strain for conjugal transfer into recipient strain, CFT073lacZYA. Transconjugants were plated on LB agar supplemented with kanamycin (to select for the transposon) and nalidixic acid (to select for the recipient strain, CFT073lacZYA). A library of these insertion mutants can then be screened for beta-galactosidase enzyme activity on either MOPS agar plates containing X-Gal (5-bromo-4-chloro-3-indoyl beta-D-galactoside) or standard MacConkey agar. On each of these medias, production of beta-galactosidase enzyme by bacterial colonies can be assayed by a color change. CFT-073lacZYA has no endogenous beta-galactosidase enzyme activity, so any detected activity is due to expression of genes near where the transposon has been inserted into the genome. Each insertion mutant can be grown on the two indicator medias in the presence or absence of 500 micrograms per milliliter D-serine. (Nelson K. M. et al., Infect. Immun,. 2001, 69(10):6201-6208.)
Among the urosepsis and urinary tract isolates tested, 49 of 60 were able to grow on D-serrine minimal medium. In contrast, only four of the 74 strains from the DEC collection of hemorrhagic colitis and infantile diarrheal strains were able to grow on D-serine minimal medium.
To evaluate the role of D-serine catabolism in uropathogenesis, a dsdA deletion mutant lacking the 445-bp corresponding to bases 18-486 of SEQ ID NO:5 was constructed as described above. The mutant, designated WAM2615, was found to lack the ability to grow on D-serine minimal medium. When DsdA was provided in trans, WAM2615 was able to grow on D-serine minimal medium. Growth of wild-type CFT073 and the dsdA mutant in glycerol minimal medium, human and mouse urine were compared. Neither wild-type CFT073 nor the dsdA mutant grew appreciably in mouse urine. The dsdA mutant grew at approximately the same rate as wild-type CFT073 in glycerol minimal medium, in which each had a doubling time of approximately 1.2 hours (
The growth defect of the mutant in urine prompted a microscopic examination of its cellular morphology. Photomicrographs of the wild-type and mutant cells corresponding to the 3 h time point of the urine growth curve showed that the dsdA mutant lost the typical rod-shape seen with the wild-type and the pleiomorphic mutant cells were rounded and swollen. After 24 h in urine, the mutant cells became more rod-like and smaller in size. The cell shape defect observed for the mutant in urine was complemented when DsdA was provided in trans. When the mutant and wild-type were grown in L broth or glycerol minimal medium they appeared as rods that were similar in size (data not shown).
The ability of dsdA mutant to establish infection was evaluated in a murine model of ascending UTI, as described above. Two days postinfection, the mutant was recovered at higher frequencies than the wild-type from both bladders and kidneys (
In order to rule out the possibility that unknown mutations in the dsdA mutant background were responsible for the competitive advantage, an intact dsdA gene was provided in trans. The complemented strain regained the ability to grow in D-serine minimal medium. For UTI experiments, the wild-type strain was transformed with the complementing plasmid in order to compensate for dsdA copy effects. When these strains were tested in the murine model, no statistically significant difference was observed in their recovery after a two day infection (
Several groups have shown that expression of type 1 fimbriae is critical for establishing a UTI in different model systems (Connell et al., PNAS, 1996, 93: 9827-9832; Mulvey et al., Science, 1998, 282: 1494-1497; Sokurenko et al., PNAS, 1998, 95: 8922-8926). To determine if alterations in type 1 fimbrial expression were responsible for the dramatic effects of the dsdA mutation in vivo, type 1 fimbriae expression was assayed. In these experiments the wild-type and dsdA mutant strains were grown as described in Experimental procedures for the UTI studies. A pair of CFT073 mutants that are locked ON or locked OFF for expression of type 1 fimbriae (gift of Dr. Harry Mobley) were used as controls. The bacteria were diluted two-fold in microtiter wells and guinea pig erythrocytes were added to each well. The CFT073 type 1 locked ON strain had agglutination titers approximately two- to four-fold higher than wild-type CFT073. As expected, the CFT073 type 1 locked OFF mutant failed to agglutinate the erythrocytes. The agglutination titer for the dsdA mutant was two-fold less than the wild-type (data not shown).
Lipopolysaccharide was isolated from the wild-type and the dsdA mutant. No differences were observed in the relative amounts or sizes of their lipopolysaccharides. In addition, no differences were observed in the expression of hemolysin (data not shown).
In order to identify genes with altered expression in response to the presence or absence of D-serine, a CFT073 library of random promoter probe miniTn10lux transposon insertions was constructed as described above. The transposon carries a promoterless luciferase gene that allows detection of loci potentially regulated by D-serine. Two thousand mutants grown on glycerol minimal medium in the presence or absence of D-serine, and observed for changes in light production using an X-ray film overlay.
The mutants demonstrating changes in light production were subsequently grown in glycerol medium liquid cultures and the levels of luminescence in the presence or absence of D-serine were quantitated. Four mutants with 3-fold or greater changes in luminescence were identified, and the sequence of the miniTn10lux insertion identified by subcloning and DNA sequence analysis. Using this approach for four mutants having increased expression of luciferase in the presence of D-serine, transposon insertions were found in the direction of transcription in coding sequence of genes. These genes were: yjiY, a MG1655 gene with weak sequence similarity with a carbon starvation protein; a MG1655 gene, manY involved in transport of mannose, ppsA (phosphoenol-pyruvate synthase), an enzyme involved in gluconeogenesis, and an orf of unknown function. Interestingly, the unknown orf appears to be inserted within known genes involved in synthesis, secretion and immunity to the microcin H47, an antibiotic peptide produced by the H47 E. coli strain. Based on the comparison of the three E. coli genomes, the microcin 47 genes are present within a unique segment of CFT073 chromosome that is a new PAI present at the aspV tRNA site.
A CFT073 lacZY deletion mutant was used as the recipient for the miniTn5lacZY transposon to generate transposon mutants. To date, 2,000 colonies have been screened for changes in LacZ phenotypes. Twenty mutants having either increased or decreased LacZ expression upon exposure to non-inhibitory levels of D-serine (50 μg/ml) in complex media after overnight growth have been identified. The analysis of these mutants is underway.
The mutants will be subjected to Southern blotting analysis to assess the degree to which the insertions occur at different genomic locations. The insertion site for the D-serine regulated locus will be identified by either inverse PCR methods or direct selection of recombinant clones with the miniTn5lacZY kanamycin resistance gene and subsequent DNA sequence analysis of the transposon-chromosomal junction site using miniTn5lacZY termini-specific primers.
Mutants in which expression of dsdC and dsdXA genes is independent of D-serine will be constructed by dsdC and dsdXA constitutive promoter mutations or by the recombinant construction of dsdC and dsdXA operons under the control of D-serine-independent, conditional promoters. D-serine deaminase constitutive mutants will be isolated as described by McFall et al. using chemical mutageneis and enrichment techniques in which D-serine is a growth limiting substrate (Bloom et al., J. Bacteriol., 1975, 121: 1078-1084; Bloom et al., J. Bacteriol., 1975, 121(3): 1092-1101; McFall, Molecular Biology, 1964, 9: 754-762; McFall, J. Bacteriol., 1967, 94(6): 1982-1988; McFall, J. Bacteriol., 1975, 121(3): 1074-1077; McFall, J. Bacteriol., 1973, 113(2): 781-785; McFall et al., Mol. Gen. Genet., 1970, 106(4): 371-377) which are incorporated by reference herein.
Alternatively, a mutant having D-serine-independent expression of the dsdCXA genes may be constructed from a mutant in which either the dsdC or dsdCXA promoter is deleted and replaced with a different promoter. Using the λ-red recombinase method, PCR-generated recombinant fragments encoding a tightly-controlled promoter such as the arabinose-inducible BAD promoter will replace the native promoter for independent control of transcription of either dsdC or dsdXA. Using this promoter will require controls that eliminate or minimize arabinose-specific effects outside of D-serine-affected sites. The native promoter-deletion mutant parent strain will be used under arabinose-inducing conditions as a control. Other suitable replacement promoters may include Class III-regulated CAP-cAMP promoter (MalT) or CAP-cAMP independent, inducible promoters. With the above constructs the expression of dsdC or dsdXA can be put under DsdC-D-serine-independent control.
Using either the constitutive promoter mutants or BAD-dependent constructs, loss of function mutations for each dsdCXA gene alone and selected combinations will be constructed. This will permit examination of a variety of phenotypes independent of D-serine-induced expression. For example, dsdA and dsdA-dsdC double mutants in these backgrounds would enable study of dsdA phenotypes that are either dsdC-dependent or -independent. In addition, in a dsdC mutant where dsdXA is under D-serine independent expression, possible D-serine transport via DsdX can be studied without the changes in DsdX expression.
Genetic screening to identify dsdC-dependent and -independent/D-serine regulated genes will involve assessing LacZ phenotypic changes in a miniTn5 lacZY CFT073 mutant library in which dsdC is under control of a conditional promoter. This strategy was used to identify RscR-regulated genes in Y. enterocolitica (Nelson, et al., 2001). However, in our system, the D-serine transporter gene, dsdx must be under transcriptional control that is independent of dsdC and D-serine. A dsdA deletion will be introduced into a dsdXA constitutive promoter mutant isolated as described above. The cellular D-serine levels can be controlled in such a mutant during in vitro growth. Alternatively, starting with a CFT073 dsdXA deletion mutant background, dsdx would be reinserted, but under control of an independent constitutive promoter. For the conditional expression of dsdC, a dsdC deletion mutant of either of the previous dsdX constructs will be complemented by a dsdC recombinant plasmid with dsdC under control of the araBAD (Pbad) promoter from vectors pBAD18 or pBAD33 (Guzman, et al., J. Bacteriol., 1995, 177(14): 4121-4130). CFT073 LacZ transcriptional fusion mutants altered in expression in the presence of arabinose and D-serine will be quantified by β-galactosidase assays. Mutants with 3-fold or greater changes in β-galactosidase activity in the presence of D-serine and arabinose will be chosen for further study. A secondary screen will identify the fusions that are independent of dsdC and controlled by D-serine alone. The dsdC-recombinant plasmid from the mutants will be cured by repeated rounds of nonselective growth as done by Nelson et al. (Nelson et al., 2001) The cured mutants will be assessed for retention of the D-serine-specific phenotype. Those mutants that are regulated by arabinose alone will be eliminated from further study. Transposon insertion sites of mutants specific for dsdC or D-serine will be identified as described above.
Protein expression of wild type CFT073 and CFT073 dsdA mutant grown in human urine and glycerol minimal medium plus or minus D-serine will be compared. Total and envelope-enriched cellular proteins will be separated by two-dimensional electrophoresis (Kendrick Labs, Madison, Wis.). Samples will be prepared in triplicate and subjected to independent electrophoretic runs. Individual polypeptide spots will be isolated. Polypeptide species with reproducible changes in expression levels in response to loss of DsdA will be identified by MALDI-TOF mass spectrometric identification of peptides by molecular weight (Kendrick Labs, Madison, Wis.).
Microarray techniques will be used to assess the patterns of genome-wide gene expression controlled by D-serine (+/−) DsdC. The initial comparative analysis will use RNA purified from CFT073 cells grown in glycerol minimal medium (+/−) D-serine. Initially studies will use the E. coli K-12 spotted orf arrays, which are available from the University of Wisconsin Gene Expression Center (GEC) (http://www.gcow.wisc.edu/gec/index.html) (Wei et al., J. Bacteriol., 2001, 183(2): 545-556). Based on preliminary data, it is expected that CFT073 cells grown in minimal glycerol medium (+/−) D-serine will show expression differences for at least dsdXA, ppsA and manY, where those genes are represented on the E. coli K-12 arrays. These genes will serve as positive controls. CFT073-specific oligomer arrays will be made at the UW Gene Expression Center by the NimbleGen technology, a maskless array synthesizer that produces 125,000 custom oligonucleotide arrays with 20 base oligomers present at 16 μm×16 μm spots.
Microhybridization will be performed according to the protocol provided by GEC. Labeled probes will be generated starting with total RNA extracted from bacteria by the method described by the Gene Expression Center. Cells harvested from liquid cultures are treated with one part boiling lysis solution (2% SDS, 16 mM EDTA, 200 mM NaCl) to two parts culture solution and boiled for 5 minutes with vortexing. This aqueous solution will then be extracted three times with hot acid-phenol-chloroform and once with chioroform-isoamyl alcohol. Total RNA will be precipitated with isopropanol overnight at −20° C. and washed once with cold 70% ethanol. RNA samples will then be resuspended in DEPC-treated water and treated with RNase free RQ1 DNase (Promega). Each reaction will be then re-purified using an RNeasy kit (QIAgen) and further precipitated with lithium chloride (Ambion) according to each Manufacturer's protocol. Samples will be resuspended in DEPC-treated water and the concentration and purity of each determined by absorbance at 260/280 nm. Integrity of the rRNA bands will be determined by running each sample on a formaldehyde-agarose gel.
Samples of total RNA will be labeled with either Cy5- or Cy3-labeled dUTP (Amersham-Pharmacia) according to the protocol supplied by GEC. Briefly, 20 μg of each sample will be mixed with 10 μg of random hexamer primers (Amersham-Pharmacia), heated to 70° C. for 5 minutes, then placed on ice. The 1X labeling mixture (containing first strand buffer [GibcoBRL], DTT, deoxynucleoside triphosphates [low dTTP, Promega], RNAsin [Promega] and DEPC water) will be added to each sample and incubated at room temperature for 10 minutes. Cy5-dUTP (or Cy3-dUTP)-labeled cDNA will be generated by reverse-transcription with Superscript II (GibcoBRL) at 42° C. for 2 hours in the dark. After labeling, remaining RNA will be hydrolyzed by the addition of NaOH and incubation for 15 minutes at 65° C. followed by the addition of HCl and Tris (pH 7.4) to neutralize the reactions. Control and experimental probes will be purified, combined, and concentrated using Microcon MY-30 filter units (Amicon).
The labeled probes will be mixed with one μg salmon sperm DNA, 4 μg of yeast tRNA and 10 μl of PerfectHyb Plus hybridization buffer (Sigma) to a final volume of 20 μl. This mixture will be heated for 5 minutes at 100° C. and hybridized to the microarray overnight (˜16 hours at 60° C.). Microarrays will then be washed 2 minutes in 0.2×SSC, 0.1% SDS, followed by two washes in 0.2×SSC, and a final wash in 0.05×SSC for 15 seconds.
Hybridization slides will be dried by centrifugation and immediately scanned using a Packard BioChip SA5000 scanning system. This system includes a scanning laser confocal fluorescence microscope and the ScanArray software that collects the 16-bit TIFF images. These images will be quantified using the QuantArray software, which automates spot finding and allows the quantitation of spot intensity.
Induction ratios for each gene will be calculated by dividing background-adjusted intensities of the experimental sample by the background-adjusted intensities of the control sample. The intensities used for these calculations will be expressed as a percentage of the total intensity of all the spots on each array, which corrects for the specific activity of probes used between arrays. Those genes having induction ratios of greater than 2 standard deviations from the mean induction ratio will be selected for further analysis and confirmation by Northern analysis.
Currently, most array data are analyzed by observing the level of induction and choosing an arbitrary cut-off point for significance. The GEC is investigating more rigorous and computationally intense methods of array data analysis. DNA microarray experiments are expensive, so replication of experiments needs to be minimized. To address this, a Bayesian statistical approach that includes prior information about the arrays (from previous experiments) in weighting the expression ratios can be used to improve the confidence of measurements (Long et aL, J. Biol. Chem., 2001, 276(23): 19937-19944). Software that applies this method, Cyber-T, is available freely from the genomics web site at the University of California at Irvine (Long et al., 2001).
Northern blotting or primer-extension analysis will be used to confirm the change in transcription levels. Candidate genes that are altered in their expression upon exposure to D-serine will be assessed for likelihood of their impact on uropathogenesis. Solid candidate genes will be mutagenized and mutants assessed for their competitive ability against the CFT073 wild type strain in the mouse UTI model.
In our laboratory, the Valdivia and Falkow DFI technique (Valdivia, et al., Science, 1997, 277: 2007-2011) permitted identification of 12 CFT073 genes that are up-expressed in the peritoneal cavity of mice but down-expressed during growth in L-broth. We constructed two separate CFT073 libraries containing 1-3 kb partial Sau3A digest fragments inserted into the promoterless GFP vector, BVC. We will reuse the libraries to identify promoters that are negatively or positively affected by growth in the presence of D-serine. This approach will identify genes that undergo relatively large changes in their expression under the alternative conditions (presence or absence of D-serine). The DFI method may also be applied with a new library constructed in a CFT073 dsdA mutant background. This new library would then be put through selections of either growth in human urine or infected mouse bladders or kidneys in the UTI model. The D-serine-positive genes that are selected and eventually identified by DNA sequence analysis will be subjected to Northern blotting analysis to confirm the induction of steady-state levels of gene-specific mRNA. Based on the criteria described above, candidate genes will be chosen for mutant construction and phenotypic analysis.
Direct binding of DsdC to candidate genes will be evaluated in the presence and absence D-serine. DsdC will be obtained from cellular extracts from strains in which DsdC is constitutively expressed or over expressed in vectors such as pT7 in a BL21 (DE3) pLysS background (Novagen). Candidate promoter fragments will be synthesized by PCR amplification, labeled internally with [α-32] dNTP or terminally by [γ-32] dNTP, and purified. The labeled promoter fragments will be incubated in binding reactions including variables as described in Ausubel et al (Ausubel et al., Current Protocols in Molecular Biology, 1987, New York: John Wiley and Sons). Sample mixtures will be separated by nondenaturing polyacrylamide gel electrophoresis and the radiolabeled bands identified by autoradiography. If warranted based on the perceived importance of the identified targets to uropathogenesis, we will purify reagent amounts of DsdC for more detailed studies. DsdC will be acquired by standard over-expression vector technology or gene fusion affinity chromatography methods. To more precisely map the DsdC binding sites, the purified DsdC (+/−D-serine) will be used in DNA mobility shift (gel-retardation) assays with overlapping PCR fragments and DNase I footprinting methods (Ausubel et al., 1987).
Binding of DsdC protein to nucleotide sequences could also be detecting by labeling the DsdC protein with a detectable label, using the labeled DsdC to probe of an oligomer array of nucleotide sequences from a UPEC strain, detecting binding of DsdC to oligomers, and identifying the oligomer. The binding assays may be performed in the presence or absence of D-serine.
In vitro transcription reactions will be performed as described previously (Aiyar et al., PNAS, 1998, 95(25): 14652-14657; Weyand et al., Mol. Microbiol., 2001, 39(6): 1504-1522), in the presence and absence of D-serine. These experiments will permit identification of newly synthesized labeled transcripts by gel electrophoresis and autoradiography. McFall's group demonstrated that D-serine deaminase activity could be detected in DsdC-dependent in vitro transcription/translation reactions using recombinant plasmids (Heincz et al., J. Bacteriol., 1984, 160(1): 42-49). This result suggests that in vitro transcriptional analysis with DsdC-dependent promoters is feasible.
To directly assess the relevance of the D-serine-regulated genes that are identified, loss-of-function, in-frame deletion mutants will be constructed utilizing the PCR fragment λ-red recombinase method (Datsenko et al., 2000) and tested for the ability to compete with CFT073 wild type in the UTI model as described above.
Intracellular invasion assays of 5637 human bladder epithelial cells as described by Mulvey et al. (Mulvey et al., Science, 1998, 282(5393): 1494-1497; Mulvey et al., Infect. Immun., 2001, 69(7): 4572-4579), which are incorporated by reference herein, will be performed to compare the abilities of the CFT073 and CFT073 dsdA strains to invade and replicate. We have acquired from Dr. Scott Hultgren, E. coli strain NU14 will be used as a positive control in the invasion studies. In addition, two CFT073 mutants that are locked “ON” or “OFF” will be included as controls for type 1 pili expression and mannose-sensitive guinea pig RBC-agglutination activity. These strains were the kind gift of Dr. Harry Mobley. The CFT073 type 1 pilus “OFF” mutant should serve as the negative control for the invasion studies. We expect that these experiments will help to assess if the 300-fold greater competitive ability of the CFT073 dsdA mutant is correlated with either increased efficiency in attachment, intracellular invasion or intracellular persistence. The latter phenotype will be judged to occur if intracellular organisms do not decrease over time or there appears to be decrease in the killing of the cells in the infected epithelial cell layers. The uroepithelial invasion assay will be used to assess the phenotypes of the additional mutants isolated as described above.
The present invention is not limited to the exemplified embodiments, but is intended to encompass all such modifications and variations as come within the scope of the following claims.